Gallaudet U just launched an app for videoconference captioning

With almost two years of video meeting experience under our belts, DC’s Gallaudet University and McLean, Virginia speech tech provider AppTek are launching a new tool for audio accessibility: a videoconference captioning app for the deaf and hard-of-hearing community.

Gallaudet offers a higher education curriculum for deaf and hard-of-hearing students via English and American Sign Language programming, while AppTek develops software in human language tech. They teamed on the app, called GoVoBo, to offer users live transcription and captions in videoconferencing, alongside other user interface tools to assist in the real-time readability of video conversations. AppTek developed the technology with Gallaudet’s Technology Access Program (TAP) and its School of Science, Technology, Accessibility, Mathematics and Public Health. Gallaudet’s Christian Vogler and Patrick Boudreault led the project at the university.

AppTek and Gallaudet first announced the technology last year, although the pair had been collaborating for years to develop research on accessibility in audio captioning. But Vogler and Boudreault told Technical.ly that with the rise of remote work and learning, automatic speech recognition (ASR) tech has been more important than ever for the community.

“Caption usage has been always in high demand and widespread, especially among deaf, hard of hearing, and late-deafened individuals for multiple reasons,” Volger and Boudreault said in emailed remarks to Technical.ly. “…Although the prevalence of auto-generated captioning prior to the pandemic is not known, it is reasonable to assume that increased usage of video conferencing calls — according to LiveWebinar (2020), 475 million people use video conferencing daily — has led to the increasing need of ASR to improve communication and understanding, review missed information and document meeting activity or information.”

According to AppTek, the GoVoBo app was built with an in-house ASR engine and web backend, both of which were developed by the tech firm. Frontend aspects, which combine audio tech, HTML, Javascript and WebRTC via the Chromium Embedded Framework, were developed by Gallaudet.

With the rise of online tools over the past year and a half, AppTek and Gallaudet said that online video conferencing continues to be a challenge for the deaf and hard-of-hearing community. For captions, many have to complete multiple, complex steps to set up the technology and manage different caption tech across multiple conferencing apps. Some video platforms don’t have ASR and captioning built-in, and third-party tech is often not interchangeable with multiple platforms. The GoVoBo app, on the other hand, is platform-agnostic and user-centered.

Another key part of the GoVoBo app, the pair said, is that because it was built by Gallaudet, it was designed directly by and for the deaf and hard-of-hearing community. It also, Vogler and Boudreault said, is not limited only to videoconferencing, and they believe it will eventually reduce licensing costs in caption tech.

“With deaf people leading the way in full partnership with AppTek by designing and implementing features carefully crafted based on deaf and hard of hearing people’s feedback, the GoVoBo platform provides a flexible and universal platform to overcome accessibility barriers for all…” said Vogler and Boudreault. “This refined approach by Gallaudet and AppTek successfully heralds a new era of telework with the integration of multimedia tools and sign language interpreting.”